Identification of the components, targets, and magnitude of an effective immune response to HIV are important steps toward the development of effective vaccines or immunotherapies. Although patients with normal CD4+ T cell counts and low levels of plasma virus are a heterogeneous group, a small subgroup of patients with truly non-progressive HIV infection and restriction of virus replication likely hold important clues to the basis of an effective immune response to HIV. A small subpopulation (fewer than 0.8% of HIV infected individuals) shows no signs of progression over a 10 year period. We have assembled a unique cohort of such patients with non-progressive disease termed long-term nonprogressors(LTNP). Many of these patients have been infected for 20 years in the absence of antiretroviral therapy with no CD4+ T cell decline, and plasma viral RNA below 50 copies per milliliter. Cells from these patients are being used to systematically dissect the mechanisms of immune mediated restriction of virus replication. HIV-specific CD8+ T cell responses of these patients have been studied by a combination of standard cytotoxic T cell (CTL) assays, flow cytometric assays, and T cell receptor (TCR) repertoire analysis in comparison with patients with progressive disease. HLA typing revealed a dramatic association between the HLA B*5701 class I allele and nonprogressive infection (88% (15 of 17) vs 9.5% (19 of 200) in progressors). The gag-specific CD8+ T cell response in the LTNP group was highly focused on peptides previously shown to be B57 restricted. These findings indicated that within this phenotypically and genotypically distinct cohort, a host immune factor is highly associated with restriction of virus replication and nonprogressive disease. They also strongly suggested a mechanism of virus specific immunity that directly operates through the B5701 molecule. Very high frequencies (0.8-18.0%) of circulating CD8+ T cells were found to be HIV specific. However, high frequencies of HIV-specific CD8+ T cells were not limited to LTNP with restriction of plasma virus. We have also observed that high frequencies of HIV-specific CD8+ T cells capable of recognizing autologous virus variants persist in LTNP and progressors. Thus, the greater ability of HIV-specific CD8+ T cells of LTNP to restrict HIV replicaiton could not be attributed to differences in CD8+ T cell frequency, specificity, ability to activate, ability to produce cytokines or ability to recognize autologous viral variants. Taken together, these data suggested that other qualitative properties of HIV-specific CD8+ T cells may distinguish the response of LTNP from those of progressors. We have recently observed that LTNP are distinguished by the maintenance of HIV-specific CD8+ T cells with a high proliferative capacity. This proliferation is coupled to or parallels perforin expression. By comparison, the HIV-specific CD8+ T cells of progressors had a limited proliferative capacity, limited ability to upregulate perforin expression, and accumulated in early G1 of the cell cycle. These results indicate that LTNP are differentiated by increased proliferative capacity of HIV-specific CD8+ T cells linked to enhanced effector function. In addition, the relative absence of these functions in progressors may represent a mechanism by which HIV avoids immunological control. Strong HIV specific CD4+ T cell proliferative responses are demonstrable in these LTNP patients. However, they are not demonstrable in the majority of infected patients with progressive disease. It has been presumed that HIV-specific CD4+ T cells are killed upon encountering antigen and maintenance of CD4+ T cell responses in some patients causes the restriction of virus replication. We have recently shown that although proliferative responses were absent in patients with poorly restricted virus replication, HIV-specific CD4+ T cells capable of producing inteerferon-gamma were detected. In two separate cohorts of 8 and 15 patients, interruption of antiretroviral therapy resulted in the rapid and complete abrogation of virus-specific proliferation although HIV-1-specific CD4+ T cells were present in the peripheral blood during viremia. Diminished HIV-specific proliferative responses were associated with diminished production of IL-2 by HIV-specific CD4+ T cells. Further, HIV-specific CD4+ T cells of viremic patients could be induced to proliferate in response to HIV antigens when co-stimulation was provided by either anti-CD28 antibody or IL-2 in vitro. In 12 of 15 patients, HIV replication was not controlled below 5,000 copies per milliliter of plasma for greater than 12 weeks. Thus, HIV-1-specific CD4+ T cells persist but do not proliferate in viremic patients. These cells retain the ability to activate, produce cytokine, and are not replicatively senescent. Taken together, these cross-sectional and longitudinal data suggest the absence of HIV-specific proliferative responses in most untreated patients does not represent a critical defect that permits poorly restricted HIV replication, but rather is a consequence of viremia. These findings raise a number of fundamental issues for the potential use of the T cell response to HIV in prophylactic vaccines or immunotherapies. Because most individuals fail to restrict HIV replication despite a broad, high frequency CD8+ T cell response, simply stimulating a high frequency response via a therapeutic or prophylactic vaccine will not necessarily result in durable immunologic control. Thus, it is unclear whether CD8+ T cells can be sufficiently primed such that qualitative changes in the CD8+ T cell response can be avoided. In addition, it is unclear how long after HIV infection of vaccinees these qualitative changes might occur and immunologic restriction of viral replication is lost. Further, it is unclear whether these qualitative changes can be reversed in infected individuals. A deeper understanding of the basis of immunologic control in LTNP and the loss of immunologic control in progressors is likely to provide information that is critical for use of the cellular immune response in each of these settings.